CN112070357A - Radar radiation source threat assessment method based on improved BP neural network - Google Patents
Radar radiation source threat assessment method based on improved BP neural network Download PDFInfo
- Publication number
- CN112070357A CN112070357A CN202010799591.9A CN202010799591A CN112070357A CN 112070357 A CN112070357 A CN 112070357A CN 202010799591 A CN202010799591 A CN 202010799591A CN 112070357 A CN112070357 A CN 112070357A
- Authority
- CN
- China
- Prior art keywords
- radiation source
- threat
- radar radiation
- neural network
- following formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
- G01S7/36—Means for anti-jamming, e.g. ECCM, i.e. electronic counter-counter measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Mathematics (AREA)
- Development Economics (AREA)
- Software Systems (AREA)
- Educational Administration (AREA)
- Strategic Management (AREA)
- Pure & Applied Mathematics (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computing Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Health & Medical Sciences (AREA)
- Radar, Positioning & Navigation (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Algebra (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Remote Sensing (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
Abstract
The invention discloses a radar radiation source threat assessment method based on an improved BP neural network, which comprises the following steps: 1. constructing a radar radiation source threat assessment model, and determining radar radiation source threat factors; 2. constructing a new weight evaluation model of radar radiation source threat factors by utilizing an analytic hierarchy process and an entropy weight process; 3. establishing membership functions of all threats; 4. calculating a threat value of a radar radiation source; 5. the accuracy rate of evaluating the radar radiation source threat is improved by improving the BP neural network. The invention utilizes the data obtained by the previous evaluation mode to compare with the training of the experience value hidden layer neuron network to obtain the optimal hidden layer neuron so as to improve the evaluation performance of the radar radiation source threat.
Description
Technical Field
The invention relates to the technical field of electronic countermeasure, in particular to a radar radiation source threat assessment method based on an improved BP neural network.
Background
The threat of a radar radiation source to our party is judged, and many factors such as carrier frequency, pulse repetition frequency, platform speed, platform distance, target type and the like need to be considered. In the face of increasingly complex electromagnetic environments and the emerging electromagnetic threats, effective assessment of battlefield radar radiation source threats is difficult to achieve.
Most methods for index weight at the present stage are from a single aspect, either the entropy of the information is objective, or the influence of the factors on the threat level is judged artificially and subjectively, and the constructed evaluation model also has certain limitations and is difficult to improve the whole evaluation performance.
The threat degree of the radar radiation source is evaluated by adopting a subjective and objective combination mode, the influence of the index which is considered to be important by human subjectivity on the threat of the radiation source is utilized, meanwhile, the data information of each index of the radar radiation source is fully utilized, and the evaluation model obtained by the mode has better representativeness.
The number of hidden layer neurons in the existing BP neural network is basically determined by empirical values, the number of hidden layer neurons affects the structure of the BP neural network, the structure of the BP neural network has great influence on the training effect of the whole network, and the method for effectively and accurately determining the number of hidden layer neurons is not provided in the prior art.
Disclosure of Invention
Aiming at the defects of the existing radar radiation source threat assessment method, the invention provides a radar radiation source threat assessment method based on an improved BP neural network, and the data obtained by using the previous assessment mode is compared with the training of the hidden layer neuron network by using empirical values to obtain the optimal hidden layer neuron so as to improve the radar radiation source threat assessment performance.
The technical scheme adopted by the invention is as follows: and constructing a new radar radiation source threat value evaluation model by re-inducing the radar radiation source threat factors, and realizing the evaluation of the radar radiation source threat level by utilizing the improved BP neural network so as to improve the accuracy of threat evaluation. The implementation steps comprise:
a radar radiation source threat assessment method based on an improved BP neural network comprises the following steps:
(1) constructing a radar radiation source threat value evaluation model and determining radar radiation source threat factors;
(2) constructing a weight evaluation model of radar radiation source threat factors by utilizing an analytic hierarchy process and an entropy weight method to obtain a weight matrix W;
(3) establishing membership functions of all threats to obtain a fuzzy comprehensive evaluation matrix R;
(4) calculating a threat value B of a radar radiation source by using a weight matrix W obtained by the analytic hierarchy process and the entropy weight method in the step (2) and a fuzzy comprehensive evaluation matrix R obtained by the step (3) based on the membership function, and taking the fuzzy comprehensive evaluation matrix R and the threat value B as an input vector and a label of the BP neural network;
(5) the accuracy of evaluating the threat value of the radar radiation source is improved by improving the BP neural network: and determining the range of the number of the hidden layer neurons by using the empirical value, evaluating by using each hidden layer neuron, comparing the error of each output with a reference value, and determining the number of the hidden layer neurons corresponding to the minimum error, namely the optimal number.
Further, the radar radiation source threat value evaluation model constructs a two-stage radar radiation source threat evaluation system from 3 aspects of radar radiation sources, platform types, motion factors, radiation source parameters and working modes, wherein the two-stage radar radiation source threat evaluation system comprises the radiation source types, the platform types, the speed, the relative distance, the motion direction, the pulse repetition frequency, the carrier frequency, the pulse width and the working modes.
Further, the step (2) comprises the following steps:
(2a) constructing judgment matrixes of all levels, and adopting a 1-9 scale theory to compare the same-layer factors pairwise to establish a judgment matrix;
(2b) comparing the radiation source type and the platform type pairwise to obtain a matrix A1:
(2c) Comparing every two of the speed, the relative distance and the motion direction to obtain a matrix A2:
(2d) Comparing the pulse repetition frequency, the carrier frequency, the pulse width and the working mode to obtain a matrix A3:
(2e) Comparing the type of the first-level index radiation source, the motion factors, the radiation source parameters and the working mode to obtain a judgment matrix A4:
(2f) And (3) judging the inconsistency degree of the paired comparison matrixes by using a consistency index CI:
CI ═ λ -n)/(n-1); wherein λ is the maximum feature root of the pairwise comparison matrix a, and a satisfies consistency when CI is 0; the larger the CI, the higher the inconsistency of A;
(2g) introducing a random consistency index RI, randomly constructing a positive reciprocal array A' of a certain order, calculating a consistency index CI thereof as a sample, and taking an average value as the random consistency index of the order matrix on the basis of a large number of samples;
(2h) calculating the relative weight vector of each index according to the following formula:
W=(w1,w2,…,wn)T
wherein the content of the first and second substances,pijthe specific gravity of the index value of the ith item under the jth index;
(2j) calculating the weight by using an entropy weight method;
(2k) calculating the specific gravity p of the index value of the ith item under the jth index according to the following formulaij:
(2l) calculating the entropy e of the jth index according toj:
Wherein k is 1/lnm;
(2m) calculating the entropy weight w of the j-th index according to the following formulaj:
(2n) determining the integrated weight of the indicator according to the following formula:
wherein alpha isj(j ═ 1,2, …, n) weights the importance of the index for the evaluator according to its own purpose and requirements;
(2o) determining a comprehensive determination weight value according to the following formula:
wherein, w1iWeight obtained for the analytic hierarchy process, w2iFor the weights obtained by the entropy weight method, the weight matrix W ═ W1 W2… Wi](i=1,2,3,…,m)。
Further, the step (3) comprises the following steps:
(3a) selecting a membership function of the radiation source type according to the following formula:
(3b) selecting a platform type threat membership function according to the following formula:
(3c) selecting a velocity threat membership function according to the following formula:
(3d) selecting a relative distance threat membership function according to the following formula:
(3e) selecting a motion direction threat membership function according to the following formula:
(3f) selecting a pulse repetition frequency threat membership function according to the following formula:
(3g) selecting a carrier frequency threat membership function according to the following formula:
(3h) selecting a pulse width threat membership function according to the following formula:
(3i) selecting a working mode threat membership function according to the following formula:
further, the step (5) comprises the following steps:
(5a) determining the basic range of the hidden layer by referring to an empirical formula, and constructing a risk assessment model according to the determined neuron numbers of the input layer, the output layer and the hidden layer, wherein the risk assessment model is constructed as t-s-u;
(5b) and (4) setting different numbers s (s is more than or equal to 4 and less than or equal to 13) of hidden layer neurons of the BP neural network, calculating the original data of the input radiation source under different s conditions to obtain an output vector, comparing the output vector with B in the step (4), and finding out the s corresponding to the minimum difference norm to serve as the number of hidden layer neurons of the BP neural network.
The invention provides a new weight evaluation model of radar radiation source threat factors by utilizing an analytic hierarchy process and an entropy weight method, establishes membership functions of each threat, further obtains a more representative radar radiation source threat value, realizes more accurate threat evaluation on a radar radiation source by improving a BP neural network and combining the BP neural network with the radar radiation source, and has the following advantages:
(1) according to the method, the threat factors are considered more comprehensively, and the weight of the threat factors is calculated by using an analytic hierarchy process and an entropy weight method, so that the weight evaluation result is closer to the reality, and the obtained threat evaluation value of the radar radiation source is more accurate;
(2) according to the invention, the BP neural network is improved, so that the node number of the BP hidden layer is more suitable for threat assessment data value training of the radar radiation source, and the finally obtained threat level assessment of the radar radiation source is more accurate.
Drawings
FIG. 1 is a flow chart of a radar radiation source threat assessment method based on an improved BP neural network;
FIG. 2 is a schematic illustration of the threat factors of the radar radiation source of the present invention;
FIG. 3 is a schematic diagram of a neural network architecture of the present invention;
FIG. 4 is a flowchart of finding the optimal number of hidden layer neurons by the improved BP neural network according to the present invention;
FIG. 5 is a graph showing the variation of the comparison error between the threat value outputted from the neural network and the threat value of the synthetic radar radiation source with the BP network node;
fig. 6 is a comparison of before and after improvement of a BP network according to the present invention with a comprehensive evaluation.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a radar radiation source threat assessment method based on an improved BP neural network, including the following steps:
(1) and constructing a radar radiation source threat value evaluation model and determining a radar radiation source threat factor.
Specifically, the radar radiation source threat value evaluation model is composed of the following 2 factors: the weight of the radar radiation source threat factors and the membership function of the radar radiation source threat factors.
And recording a weight vector formed by each threat factor to the threat level weight as w ═ w1,w2,…,wn]And satisfy
The membership function of the radar radiation source threat factors reflects the influence degree of each factor on the analysis of the threat degree, and the membership function of the threat factors forms a threat membership vector F ═ F1,F2,…,Fn]And n is the category of the membership function.
The embodiment of the invention takes threat factors into all-round consideration, and constructs a brand-new two-stage radar radiation source threat assessment system from 3 aspects of radar radiation source and platform type, motion factors, radiation source parameters and working mode, wherein the system mainly comprises the radiation source type, platform type, speed, relative distance, motion direction, pulse repetition frequency, carrier frequency, pulse width and working mode, and is shown in figure 2.
The calculation formula of the threat value of the radar radiation source is as follows:
(2) A weight evaluation model of radar radiation source threat factors is constructed by utilizing an analytic hierarchy process and an entropy weight process to obtain a weight matrix W, and the method comprises the following specific steps:
(2a) constructing judgment matrixes of all levels, and adopting a 1-9 scale theory to compare the same-layer factors pairwise to establish a judgment matrix;
(2b) comparing the radiation source type and the platform type pairwise to obtain a matrix A1:
(2c) Comparing every two of the speed, the relative distance and the motion direction to obtain a matrix A2:
(2d) Comparing the pulse repetition frequency, the carrier frequency, the pulse width and the working mode to obtain a matrix A3:
(2e) Comparing the type of the first-level index radiation source, the motion factors, the radiation source parameters and the working mode to obtain a judgment matrix A4:
(2f) And (3) judging the inconsistency degree of the paired comparison matrixes by using a consistency index CI:
CI ═ λ -n)/(n-1); wherein λ is the maximum feature root of the pairwise comparison matrix a, and a satisfies consistency when CI is 0; the larger the CI, the higher the inconsistency of A;
(2g) introducing a random consistency index RI, randomly constructing a positive reciprocal array A' of a certain order, calculating a consistency index CI thereof as a sample, and taking an average value as the random consistency index of the order matrix on the basis of a large number of samples;
(2h) calculating the relative weight vector of each index according to the following formula:
W=(w1,w2,…,wn)T
wherein the content of the first and second substances,pijthe specific gravity of the index value of the ith item under the jth index;
(2j) calculating the weight by using an entropy weight method;
(2k) calculating the specific gravity p of the index value of the ith item under the jth index according to the following formulaij:
(2l) calculating the entropy e of the jth index according toj:
Wherein k is 1/lnm;
(2m) calculating the entropy weight w of the j-th index according to the following formulaj:
(2n) determining the integrated weight of the indicator according to the following formula:
wherein alpha isj(j ═ 1,2, …, n) weights the importance of the index for the evaluator according to its own purpose and requirements;
(2o) determining a comprehensive determination weight value according to the following formula:
wherein, w1iWeight obtained for the analytic hierarchy process, w2iFor the weights obtained by the entropy weight method, the weight matrix W ═ W1 W2… Wi](i=1,2,3,…,m)。
(3) Establishing membership functions of all threats to obtain a fuzzy comprehensive evaluation matrix R; the specific steps are as follows:
(3a) selecting a membership function of the radiation source type according to the following formula:
(3b) selecting a platform type threat membership function according to the following formula:
(3c) selecting a velocity threat membership function according to the following formula:
(3d) selecting a relative distance threat membership function according to the following formula:
(3e) selecting a motion direction threat membership function according to the following formula:
(3f) selecting a pulse repetition frequency threat membership function according to the following formula:
(3h) selecting a pulse width threat membership function according to the following formula:
(3i) selecting a working mode threat membership function according to the following formula:
(4) and (3) calculating a threat value B of a radar radiation source by using a weight matrix W obtained by the analytic hierarchy process and the entropy weight process in the step (2) and a fuzzy comprehensive evaluation matrix R obtained based on the membership function in the step (3), and taking the fuzzy comprehensive evaluation matrix R and the threat value B as an input vector and a label of the BP neural network. The method comprises the following specific steps:
firstly, normalizing the numerical values of all threat factors according to membership functions of different threat factors to be used as vectors for representing threats of different radar radiation sources to obtain a fuzzy comprehensive evaluation vector R;
secondly, determining fuzzy first-level comprehensive evaluation B of certain radari=Wi T*RiWherein W isi TObtaining a weight matrix for an analytic hierarchy process and an entropy weight process;
and finally, taking the evaluation vector R and the comprehensive evaluation B as an input vector and a training label of the BP neural network.
(5) The accuracy of evaluating the threat value of the radar radiation source is improved by improving the BP neural network: and determining the range of the number of the hidden layer neurons by using the empirical value, evaluating by using each hidden layer neuron, comparing the error of each output with a reference value, and determining the number of the hidden layer neurons corresponding to the minimum error, namely the optimal number. The method comprises the following specific steps:
(5a) determining the basic range of the hidden layer by referring to an empirical formula, and constructing a risk assessment model according to the determined neuron numbers of the input layer, the output layer and the hidden layer, wherein the risk assessment model is constructed as t-s-u;
(5b) setting different numbers s (s is more than or equal to 4 and less than or equal to 13) of hidden layer neurons of the BP neural network, calculating original data of input radiation sources under different conditions of s to obtain output vectors, comparing the output vectors with B in the step (4), finding out s corresponding to the minimum difference norm as the number of hidden layer neurons of the BP neural network, and the flow is shown in FIG. 4;
(5c) training the BP with training sample data. The method comprises the steps of setting functions and parameters of a model in a training process, enabling an implicit layer transfer function to be tandig, enabling an output layer transfer function to be purelin, directly using a train function, and enabling a learning function to be learngdm. Setting parameters of a learning rate, the maximum training times epochs of the model and target precision coarse; specific parameters are shown in table 1.
TABLE 1 BP neural network Key parameter settings
(5d) And testing the collected data by using the trained neural network, and comparing the output result with the previous expected result, wherein the higher the consistency between the two results is, the better the prediction accuracy of the neural network is.
The effects of the present invention can be further illustrated by the following experiments:
1) the experimental conditions are as follows:
the experimental environment is as follows: intel Core i7CPU 2.00Ghz, 16GB memory, Windows system, Matlab.
2) Experimental contents and results:
experiment 1: the input data of the BP neural network are generated, and the results are shown in tables 2 and 3.
Table 2 raw data of radiation source
Radar | Radiation source | 1 | |
|
|
|
Type of |
1 | 1 | 0.5 | 0.5 | 0.1 | |
Type of |
1 | 0.9 | 0.9 | 0.7 | 0.5 | |
Speed of |
3 | 2 | 0.0294 | 0.6 | 0.02 | |
|
3 | 20 | 80 | 70 | 150 | |
Direction of |
0 | 10 | 6-60 | 60 | 140 | |
|
2 | 1 | 0.6 | 0.9 | 0.2 | |
|
10 | 3 | 0.5 | 4 | 0.3 | |
|
1 | 3 | 3 | 2.5 | 9 | |
Mode of |
1 | 0.8 | 0.8 | 0.8 | 0.2 |
TABLE 3 quantized data
As shown in tables 2 and 3, the training test data set is generated by simulating with reference to the measured data of the radar radiation source based on expert experience and by using a fuzzy comprehensive evaluation method based on an analytic hierarchy process and an entropy weight method, and is used for researching a radar radiation source threat evaluation method based on a neural network.
Experiment 2: the influence of different hidden layer node numbers of the BP neural network on the estimation of the threat value is shown in a figure 5.
Through the error comparison shown in fig. 5, it can be seen that the error value between the evaluation value of the BP neural network and the evaluation value of the synthetic radar radiation source threat value is minimum when the number of the neurons is 8, which indicates that when the number of the neurons is 8, the prediction effect of the network is the best, i.e., it can be determined that the network structure of the BP neural network is 9-8-1.
Experiment 3: the influence of the BP neural network on the threat value estimation before and after improvement is shown in fig. 6.
In the simulation experiment, the number of hidden layer neurons is 4-13, the neural network corresponding to each hidden layer neuron runs for 100 times respectively, the average value of the threat values of the radiation source is calculated, and the output threat value of the BP neural network is obtained by using the number of the hidden layer neurons as 7. By utilizing the improved neural network structure 9-8-1, the radiation source threat value of the improved BP neural network is obtained after 100 times of operation, and the radiation source threat values of the two methods are compared with the threat value of comprehensive evaluation.
The threat value table shows that the output threat value of the improved BP neural network is closer to the threat value of comprehensive evaluation, and the comparison graph before and after improvement shows that the closer the output broken line of the improved BP neural network is to the comprehensive evaluation broken line, and the simulation experiment contents are integrated, so that the radar radiation source threat evaluation of the improved BP neural network is more accurate and effective than the existing BP neural network threat evaluation.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Finally, it should be noted that: the above examples are only for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above examples, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (5)
1. A radar radiation source threat assessment method based on an improved BP neural network is characterized by comprising the following steps:
(1) constructing a radar radiation source threat value evaluation model, and determining radar radiation source threat factors;
(2) constructing a weight evaluation model of radar radiation source threat factors by utilizing an analytic hierarchy process and an entropy weight method to obtain a weight matrix W;
(3) establishing membership functions of all threats to obtain a fuzzy comprehensive evaluation matrix R;
(4) calculating a threat value B of a radar radiation source by using a weight matrix W obtained by the analytic hierarchy process and the entropy weight method in the step (2) and a fuzzy comprehensive evaluation matrix R obtained by the step (3) based on the membership function, and taking the fuzzy comprehensive evaluation matrix R and the threat value B as an input vector and a label of the BP neural network;
(5) the accuracy of evaluating the threat value of the radar radiation source is improved by improving the BP neural network: and determining the range of the number of the hidden layer neurons by using the empirical value, evaluating by using each hidden layer neuron, comparing the error of each output with a reference value, and obtaining the optimal number, wherein the number of the hidden layer neurons corresponding to the minimum error is the optimal number.
2. The improved BP neural network-based radar radiation source threat assessment method of claim 1, wherein: the radar radiation source threat value evaluation model constructs a two-stage radar radiation source threat evaluation system from 3 aspects of radar radiation sources, platform types, motion factors, radiation source parameters and working modes, wherein the two-stage radar radiation source threat evaluation system comprises a radiation source type, a platform type, speed, relative distance, a motion direction, pulse repetition frequency, carrier frequency, pulse width and a working mode.
3. The improved BP neural network-based radar radiation source threat assessment method of claim 1, wherein: the step (2) comprises the following steps:
(2a) constructing judgment matrixes of all levels, and adopting a 1-9 scale theory to compare the same-layer factors pairwise to establish a judgment matrix;
(2b) comparing the radiation source type and the platform type pairwise to obtain a matrix A1:
(2c) Comparing every two of the speed, the relative distance and the motion direction to obtain a matrix A2:
(2d) Comparing the pulse repetition frequency, the carrier frequency, the pulse width and the working mode to obtain a matrix A3:
(2e) Comparing the type, motion factor, radiation source parameter and working mode of the first-level index radiation source to obtain a judgment matrix A4:
(2f) And (3) judging the inconsistency degree of the paired comparison matrixes by using a consistency index CI:
CI ═ λ -n)/(n-1); wherein λ is the maximum feature root of the pairwise comparison matrix a, and a satisfies consistency when CI is 0; the larger the CI, the higher the inconsistency of A;
(2g) introducing a random consistency index RI, randomly constructing a positive reciprocal array A' of a certain order, calculating a consistency index CI thereof as a sample, and taking an average value as the random consistency index of the order matrix on the basis of a large number of samples;
(2h) calculating the relative weight vector of each index according to the following formula:
W=(w1,w2,…,wn)T
wherein the content of the first and second substances,pijthe specific gravity of the index value of the ith item under the jth index;
(2j) calculating the weight by using an entropy weight method;
(2k) calculating the specific gravity p of the index value of the ith item under the jth index according to the following formulaij:
(2l) calculating the entropy e of the jth index according toj:
Wherein k is 1/lnm;
(2m) calculating the entropy weight w of the j-th index according to the following formulaj:
(2n) determining the integrated weight of the indicator according to the following formula:
wherein alpha isj(j ═ 1,2, …, n) weights the importance of the index for the evaluator according to its own purpose and requirements;
(2o) determining a comprehensive determination weight value according to the following formula:
4. The improved BP neural network-based radar radiation source threat assessment method of claim 1, wherein: the step (3) comprises the following steps:
(3a) selecting a membership function of the radiation source type according to the following formula:
(3b) selecting a platform type threat membership function according to the following formula:
(3c) selecting a velocity threat membership function according to the following formula:
(3d) selecting a relative distance threat membership function according to the following formula:
(3e) selecting a motion direction threat membership function according to the following formula:
(3f) selecting a pulse repetition frequency threat membership function according to the following formula:
(3g) selecting a carrier frequency threat membership function according to the following formula:
(3h) selecting a pulse width threat membership function according to the following formula:
(3i) selecting a working mode threat membership function according to the following formula:
5. the improved BP neural network-based radar radiation source threat assessment method of claim 1, wherein: the step (5) comprises the following steps:
(5a) determining the basic range of the hidden layer by referring to an empirical formula, and constructing a risk assessment model according to the determined neuron numbers of the input layer, the output layer and the hidden layer, wherein the risk assessment model is constructed as t-s-u;
(5b) and (4) setting different numbers s (s is more than or equal to 4 and less than or equal to 13) of neurons in the hidden layer of the BP neural network, calculating the original data of the input radiation source under different conditions to obtain an output vector, comparing the output vector with the B in the step (4), and finding out the s corresponding to the minimum difference norm to serve as the number of neurons in the hidden layer of the BP neural network.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010799591.9A CN112070357A (en) | 2020-08-11 | 2020-08-11 | Radar radiation source threat assessment method based on improved BP neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010799591.9A CN112070357A (en) | 2020-08-11 | 2020-08-11 | Radar radiation source threat assessment method based on improved BP neural network |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112070357A true CN112070357A (en) | 2020-12-11 |
Family
ID=73661090
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010799591.9A Pending CN112070357A (en) | 2020-08-11 | 2020-08-11 | Radar radiation source threat assessment method based on improved BP neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112070357A (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113283527A (en) * | 2021-06-07 | 2021-08-20 | 哈尔滨工程大学 | Radar threat assessment method based on level indexes |
CN114217142A (en) * | 2021-09-22 | 2022-03-22 | 中国人民解放军空军工程大学 | Battlefield electromagnetic environment complexity evaluation method |
CN114580884A (en) * | 2022-02-28 | 2022-06-03 | 中国人民解放军空军预警学院雷达士官学校 | Aerospace target threat assessment method based on multi-characteristic hierarchical structure and self-adaptive approximation |
CN114637962A (en) * | 2022-05-18 | 2022-06-17 | 自然资源部第一海洋研究所 | Ocean numerical prediction product verification method and system, electronic equipment and storage medium |
CN114741876A (en) * | 2022-04-12 | 2022-07-12 | 北京建筑大学 | Intelligent inspection method for tower crane |
CN115078850A (en) * | 2022-07-19 | 2022-09-20 | 沸蓝建设咨询有限公司 | Communication base station radiation detection system |
CN115166678A (en) * | 2022-09-06 | 2022-10-11 | 中国人民解放军海军工程大学 | Complex-scene radar radiation source pulse signal sorting and cognitive countermeasure system |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650250A (en) * | 2016-12-14 | 2017-05-10 | 中国人民解放军军械工程学院 | Infrared multi-target threat sequencing method based on equivalent evaluation factors |
CN108038300A (en) * | 2017-12-07 | 2018-05-15 | 长春理工大学 | Optical fiber state evaluating method based on improved membership function combination neutral net |
-
2020
- 2020-08-11 CN CN202010799591.9A patent/CN112070357A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106650250A (en) * | 2016-12-14 | 2017-05-10 | 中国人民解放军军械工程学院 | Infrared multi-target threat sequencing method based on equivalent evaluation factors |
CN108038300A (en) * | 2017-12-07 | 2018-05-15 | 长春理工大学 | Optical fiber state evaluating method based on improved membership function combination neutral net |
Non-Patent Citations (5)
Title |
---|
夏雄等: "基于模糊神经网络的气体密封故障诊断", 《润滑与密封》 * |
孔尚萍等: "基于AHP与熵权法的空中目标威胁评估方法", 《战术导弹技术》 * |
尹健: "基于回归神经网络的雷达辐射源识别算法", 《电讯技术》 * |
朱冠兰等: "辐射源威胁等级评估技术研究", 《现代电子技术》 * |
牛海等: "基于人工神经网络的辐射源威胁评估方法研究", 《现代防御技术》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113283527A (en) * | 2021-06-07 | 2021-08-20 | 哈尔滨工程大学 | Radar threat assessment method based on level indexes |
CN113283527B (en) * | 2021-06-07 | 2022-04-29 | 哈尔滨工程大学 | Radar threat assessment method based on level indexes |
CN114217142A (en) * | 2021-09-22 | 2022-03-22 | 中国人民解放军空军工程大学 | Battlefield electromagnetic environment complexity evaluation method |
CN114580884A (en) * | 2022-02-28 | 2022-06-03 | 中国人民解放军空军预警学院雷达士官学校 | Aerospace target threat assessment method based on multi-characteristic hierarchical structure and self-adaptive approximation |
CN114580884B (en) * | 2022-02-28 | 2023-08-08 | 中国人民解放军空军预警学院雷达士官学校 | Multi-characteristic hierarchical structure and adaptive approximation space-sky target threat assessment method |
CN114741876A (en) * | 2022-04-12 | 2022-07-12 | 北京建筑大学 | Intelligent inspection method for tower crane |
CN114741876B (en) * | 2022-04-12 | 2024-03-19 | 北京建筑大学 | Intelligent inspection method for tower crane |
CN114637962A (en) * | 2022-05-18 | 2022-06-17 | 自然资源部第一海洋研究所 | Ocean numerical prediction product verification method and system, electronic equipment and storage medium |
CN115078850A (en) * | 2022-07-19 | 2022-09-20 | 沸蓝建设咨询有限公司 | Communication base station radiation detection system |
CN115166678A (en) * | 2022-09-06 | 2022-10-11 | 中国人民解放军海军工程大学 | Complex-scene radar radiation source pulse signal sorting and cognitive countermeasure system |
CN115166678B (en) * | 2022-09-06 | 2022-12-20 | 中国人民解放军海军工程大学 | Complex-scene radar radiation source pulse signal sorting and cognitive countermeasure system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112070357A (en) | Radar radiation source threat assessment method based on improved BP neural network | |
Le Son et al. | Remaining useful life estimation based on stochastic deterioration models: A comparative study | |
CN110275148B (en) | Sea clutter amplitude distribution parameter estimation method and system | |
CN112232833A (en) | Lost member customer group data prediction method, model training method and model training device | |
CN112307677A (en) | Power grid oscillation mode evaluation and safety active early warning method based on deep learning | |
CN114565021A (en) | Financial asset pricing method, system and storage medium based on quantum circulation neural network | |
CN110533109A (en) | A kind of storage spraying production monitoring data and characteristic analysis method and its device | |
CN114679310A (en) | Network information security detection method | |
CN111832748A (en) | Electronic nose width learning method for performing regression prediction on concentration of mixed gas | |
CN112034139A (en) | Method and device for judging rock burst tendency grade and electronic equipment | |
Zhang et al. | A novel framework of fuzzy rule interpolation for takagi-sugeno-kang inference systems | |
CN114124456B (en) | Safety detection method and device for comprehensive energy system, electronic equipment and storage medium | |
Chencho et al. | Structural damage quantification using ensemble‐based extremely randomised trees and impulse response functions | |
CN115508206A (en) | Method for inverting probability of rock strength parameter of jointed rock slope | |
CN115049026A (en) | Regression analysis method of space non-stationarity relation based on GSNNR | |
CN111062118B (en) | Multilayer soft measurement modeling system and method based on neural network prediction layering | |
CN114527434A (en) | Convolutional neural network-based radar interference effect evaluation method | |
CN113780346A (en) | Method and system for adjusting prior constraint classifier and readable storage medium | |
Ünal | Nonlinear chaotic analysis of usd/try and eur/try exchange rates | |
CN112348055A (en) | Clustering evaluation measurement method, system, device and storage medium | |
Karcı et al. | Derin öğrenme yöntemleri kullanılarak deprem tahmini gerçekleştirilmesi | |
CN111461558B (en) | Key data safety destruction evaluation method | |
Yasin et al. | Statistical Inferences for Multivariate Generalized Gamma Regression Model | |
CN113553754A (en) | Memory, fire risk prediction model construction method, system and device | |
Mobasher-Kashani et al. | Nonlinear regression in tax evasion with uncertainty: a variational approach |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |